Informatics Research Seminar: Using Natural Language Processing to Evaluate Electronic Health Record Documentation of Hypertension Treatment
February 20 @ 4:00 pm - 5:00 pm EST
Speaker: Kim Shoenbill, MD, MS
Presented from UNC-CH
Broadcast Link: Seminar
Over 45% of the 85.7 million US adults with hypertension have uncontrolled blood pressure resulting in increased risks of cardiovascular disease including stroke, heart failure, and myocardial infarction. Guidelines on hypertension management include lifestyle modification (e.g., diet, exercise) and medication initiation as first line treatment. To understand current hypertension treatment efforts and improve hypertension control, it is important to determine the frequency and inter-relatedness of lifestyle modification and hypertension medication initiation. However, lifestyle modification data is documented in narrative form within the electronic health record, making it “invisible” in evaluation of discrete data or metric measurement of hypertension treatment. Electronic health record data from 14,860 adult hypertension patients at an academic medical center were analyzed using natural language processing and statistical methods to determine documentation of lifestyle modification (i.e., advice and/or assessment) and hypertension medication initiation. Methods and results from this analysis will be discussed.
Kimberly Shoenbill, MD, PhD is a physician and informatician at the University of North Carolina – Chapel Hill. She is an Assistant Professor working in the Department of Family Medicine and the Program on Health and Clinical Informatics. She received her MD and her PhD in Clinical Investigation with an emphasis in Informatics from the University of Wisconsin – Madison. She is dually board certified in Family Medicine and Clinical Informatics. Her research focuses on secondary use of electronic health record data using natural language processing and machine learning coupled with statistical analysis. She is committed to using informatics to evaluate, inform, and improve patient care delivery and outcomes.